The activation of O2 by diiron cluster-containing enzymes for insertion into biological molecules usually requires a partnership between the cluster, which binds and activates O2, and the protein, which directs and regulates the process in a variety of ways. The well-studied diiron monooxygenases all use carboxylate-rich diiron cluster ligand structures with one His ligand per iron and a 4-helix-bundle protein fold. Moreover, most catalyze reactions such as C-H or aromatic hydroxylation, which are mechanistically related reactions. In order to resolve the roles of the diiron cluster and protein structure in O2 activation, and to further explore the range of chemistry accessible by the diiron cluster oxygenases, new classes of diiron oxygenases are needed that: (i) present the diiron core in a different protein environment, (ii) invoke new core structures, and/or (iii) catalyze novel reactions. While studying the biosynthetic pathway for chloramphenicol from Streptomyces, we found two new diiron enzymes that address these requirements. The first enzyme, CmlA, is the founding member of a class of at least 50 uncharacterized enzymes that catalyze essential ?-hydroxylation of antibiotics, biostatics and chemotherapy agents, as they are synthesized in nonribosomal peptide synthetase (NRPS)-based pathways. Sequence homology, spectroscopy, and product analysis show that CmlA is the first diiron monooxygenase recognized to: (i) bind the diiron cluster in a novel ?-lactamase fold, (ii) catalyze b- hydroxylation, and (iii) incorporae more than one His ligand per Fe. We have shown in preliminary studies using a variety of spectroscopies that the Fe(II)Fe(II) state of CmlA reacts with O2 only when it is complexed with its NRPS (CmlP) covalently loaded with the chloramphenicol precursor L-p-NH2-phenylalanine (PAPA) on its thiolation domain. The second new diiron enzyme, CmlI, catalyzes the final step in chloramphenicol biosynthesis, aromatic amine to nitro conversion. This chemistry has only recently been recognized in the diiron family and is poorly characterized. Preliminary studies show that CmlI uses the 4-helix bundle fold, but, like CmlA, it has more than one His ligand per Fe. Unlike CmlA, it will not catalyze any of the common diiron monooxygenase reactions, and it will accept its substrate either free or bound to CmlP. We propose to structurally characterize CmlA and CmlI with and without their substrates. Efficient single turnover systems for both enzymes will allow transient kinetic techniques to be used to search for and trap reaction cycle intermediates for spectroscopic characterization. Alternative substrates and active site mutagenesis will be used to probe the molecular mechanisms of both enzymes. Spectroscopic labels will be used to reveal the interaction zones with CmlP to investigate the means by which this component regulates O2 binding. Both the intermediates and modes of regulation will be directly compared with those of methane monooxygenase, long studied in this laboratory. This work will bear on fundamental mechanisms of O2 activation and regulation as well as potentially leading to important insights into strategies for the production of novel antibiotics.